Waste lithium battery recycling method based on multi-stage crushing and intelligent sorting

Abstract

The invention relates to the technical field of lithium battery recovery, in particular to a waste lithium battery recovery method based on multistage crushing and intelligent sorting. Performing soaking discharge treatment on the waste lithium battery, and constructing a discharge detection model; then carrying out multi-stage crushing on the discharged lithium battery by adopting first-stage shearing crushing, second-stage impact crushing and third-stage grinding crushing, and grading and screening crushed particles; an electrode material sorting model is constructed by acquiring spectrum and element information of a crushed material and combining a deep learning model, so that intelligent sorting of an electrode material and a non-electrode material is realized; and after separation, performing multi-frequency vibration layering on the non-electrode material particles, and further separating positive and negative electrode materials from the electrode material particles by adopting an electrostatic separation technology. According to the method, multiple physical sorting modes and an artificial intelligence technology are comprehensively utilized, and efficient and accurate recycling and resource utilization of the waste lithium batteries are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery recycling, and in particular to a waste lithium battery recycling method based on multi-stage crushing and intelligent sorting. Background Art

[0002] The recycling of used lithium batteries faces many technical challenges. First, different types and models of lithium batteries are discharged in a mixed manner, and their structural materials vary greatly, which increases the difficulty of sorting. Secondly, lithium batteries themselves contain flammable and toxic substances, such as electrolytes, and improper disassembly and handling can easily lead to safety accidents. In addition, the positive and negative electrode materials of lithium batteries are mixed with other components such as current collectors and diaphragms. Traditional physical sorting methods are difficult to achieve accurate separation, resulting in low purity of the recycled materials.

[0003] At present, the recycling methods of waste lithium batteries used in industry mainly include pretreatment, crushing, sorting, leaching, extraction and other steps. In the sorting process, commonly used methods include gravity, magnetic separation, flotation, etc., but the sorting efficiency and accuracy need to be improved. In addition, most recycling processes lack intelligent means, making it difficult to effectively deal with the increasingly complex composition of waste batteries, and the level of automation is not high.

[0004] In order to improve the efficiency and quality of recycling of waste lithium batteries, it is urgent to develop new intelligent and refined process routes. In recent years, artificial intelligence technologies such as machine vision, multi-sensor information fusion, and deep learning have developed rapidly, providing new ideas for realizing efficient and intelligent recycling of waste lithium batteries. Through multi-source information perception such as spectroscopy and imaging, combined with deep learning models, the type and material composition of lithium batteries can be accurately identified, thereby guiding precise disassembly and sorting. At the same time, the introduction of intelligent control and parameter optimization in the crushing, screening, and transportation links can significantly improve the sorting effect and increase the material recovery rate. In addition, the intelligentization of the recycling process helps to reduce energy consumption, reduce pollution emissions and safety hazards, and achieve green environmental protection and sustainable development.

[0005] The traditional waste lithium battery recycling process lacks automation and intelligence, and the sorting efficiency and accuracy are limited, making it difficult to meet the growing demand for lithium battery recycling.

[0006] In view of this, the present invention proposes a waste lithium battery recycling method based on multi-stage crushing and intelligent sorting. Summary of the invention

[0007] To achieve the above object, the present invention provides a waste lithium battery recycling method based on multi-stage crushing and intelligent sorting, and the specific technical scheme is as follows:

[0008] The recycled waste lithium batteries are subjected to immersion discharge treatment, and a waste lithium battery discharge detection model is constructed to detect whether the waste lithium batteries have completed discharge;

[0009] The lithium battery after discharge treatment is crushed in multiple stages, including primary shear crushing, secondary impact crushing and tertiary grinding crushing. In the process of graded crushing, the particle size of lithium battery is graded and sieved;

[0010] Obtain the spectrum and element information of the crushed lithium battery materials, build an electrode material sorting model based on the deep learning model, identify the electrode materials based on the spectrum and element information of the crushed lithium battery materials, and control the pneumatic sorter to sort the electrode materials and non-electrode materials;

[0011] Separately collect the lithium battery electrode material particles and non-electrode material particles after pneumatic sorting, apply multi-frequency vibration to the collected non-electrode material particles, and stratify the non-electrode material particles;

[0012] The collected electrode material particles are subjected to electrostatic separation to separate the positive electrode material and the negative electrode material in the electrode material particles.

[0013] Preferably, the waste lithium battery is immersed in a salt solution tank for discharge, and a current sensor is connected to the total output circuit of the battery to monitor the change of the total output current I(t) in real time. The waste lithium battery discharge detection model is:

[0014]

[0015] Where I(t) represents the total output current of the battery pack at time t, n is the number of used lithium batteries, and I i (t) represents the output current of the ith battery at time t, U i represents the initial voltage of the ith battery, i.e., the open circuit voltage before immersion discharge, R i represents the internal resistance of the ith battery, t represents the immersion discharge time, τ i represents the time constant of the ith battery, τ i =R i ×C i , where C i is the capacity of the i-th battery;

[0016] The total output current I(t) of the battery pack is the superposition of the output currents of all batteries. By detecting the size of I(t), the discharge process of the waste lithium battery can be judged;

[0017] Set the discharge current threshold I th , when the measured total output current I(t) is less than or equal to I th When , it is determined that all batteries have been discharged;

[0018] After the waste lithium battery is completely discharged, the lithium battery soaked in the salt solution is rinsed with water to remove the residual salt solution and impurities on the surface of the lithium battery, and then dried.

[0019] Preferably, the waste lithium batteries after the discharge treatment are crushed, and the crushing process includes primary shear crushing, secondary impact crushing and tertiary grinding crushing;

[0020] Introducing adaptive feeding control of waste lithium batteries based on machine vision, using convolutional neural network (CNN) to extract features and segment semantics of images in the crushing chamber, monitoring the material level of waste lithium batteries in real time, and adjusting the feeding rate by real-time detection of the material level in the crushing chamber;

[0021] The waste lithium batteries are subjected to primary shearing and crushing treatment. A temperature sensor is set in the shearing and crushing chamber of the shearing and crushing treatment, and the temperature threshold T of the shearing and crushing chamber is set. δ When the temperature T in the shear crushing chamber is detected to be greater than the temperature threshold T δ When the shearing and crushing is stopped, the shearing and crushing chamber is sprayed with water for cooling. After the water spraying and cooling is stopped, it is detected whether the temperature T in the shearing and crushing chamber is less than the temperature threshold T δ , if T <T δ , then continue shearing and crushing; if T ≥ T δ , then the high temperature alarm is triggered;

[0022] Setting the particle size threshold D of waste lithium batteries during primary shear crushing p1 , and according to the crushing particle size threshold D p1 Set the first-level screen mesh diameter d1, D p1 =d1; the crushed waste lithium batteries that cannot pass through the first-level screen are sheared and crushed again.

[0023] Preferably, after the waste lithium batteries have completed primary shear crushing, the crushed waste lithium batteries are subjected to secondary impact crushing;

[0024] Before impact crushing the broken waste lithium batteries, nitrogen is input into the impact crushing chamber to replace the air in the impact crushing chamber by the input nitrogen until the oxygen concentration in the impact crushing chamber is lower than the oxygen safety concentration threshold, an oxygen analyzer is arranged at the exhaust port of the impact crushing chamber to detect the oxygen concentration in the impact crushing chamber, and when it is detected that the oxygen concentration in the impact crushing chamber is lower than the oxygen safety concentration threshold, the broken waste lithium batteries after primary shear crushing are put into the impact crushing chamber for secondary impact crushing;

[0025] During the impact crushing process, nitrogen is continuously input to maintain the positive nitrogen pressure in the impact crushing chamber, and the oxygen concentration in the impact crushing chamber is continuously detected. If the oxygen concentration is greater than the oxygen safety concentration threshold, the impact crushing is stopped;

[0026] The nitrogen discharged during the crushing process is dried, dust-filtered and reused;

[0027] Setting the particle size threshold D of waste lithium batteries during secondary impact crushing p2 , and according to the crushing particle size threshold D p2 Set up the secondary screen mesh diameter d2, D p2 =d2; re-impact and crush the broken waste lithium batteries that cannot pass through the secondary screen;

[0028] The lithium battery after impact crushing is subjected to three-stage grinding and crushing to grind the crushed lithium battery into powder particles. The crushing particle size threshold D of the waste lithium battery during the three-stage grinding and crushing is set. p3 , and according to the crushing particle size threshold D p3 Set up a three-level screen mesh diameter d3, D p3 =d3; the crushed waste lithium batteries that cannot pass through the third-level screen are re-grinded and crushed until all the ground particles of the waste lithium batteries pass through the third-level screen.

[0029] Preferably, an X-ray fluorescence spectrometer and a laser induced breakdown spectrometer are used to collect spectral data and element information data of N different types of lithium battery materials, M samples are collected for each type of lithium battery, a total of N×M samples, and the lithium battery material samples are labeled into K material categories, with labels y i ∈1,2,...,K; denoise and normalize the spectral data;

[0030] Perform PCA feature extraction on the spectral data of lithium battery materials, and set the normalized spectral data matrix as The principal component matrix obtained by PCA is Then: X=U∑V T ; Where P is the number of principal components selected by PCA, ∑ is the singular value matrix, V is the right singular matrix, V T is the transpose of the right singular matrix; the spectral feature matrix after dimension reduction is:

[0031] Perform ICA feature extraction on the element information data of lithium battery materials, assuming that the element information matrix is The independent component matrix obtained by ICA is Then: Y = AS; where A is the mixing matrix, S is the independent component, D is the number of element types, and Q is the number of selected independent components; the element feature matrix after dimensionality reduction is:

[0032] Combine the spectral feature matrix and the element feature matrix to convert the spectral feature matrix X pca and element feature matrix Y ica Splice in the column direction to obtain a (P+Q)-dimensional sample feature matrix

[0033] Preferably, an electrode material sorting model based on a convolutional neural network (CNN) model is constructed, the input layer is a (P+Q)-dimensional sample feature matrix, after C convolutional layers, G pooling layers and B fully connected layers, the output is a K-dimensional material category probability vector

[0034] In the CNN model, let the output feature map of the l-1th layer be H (l-1) , the convolution kernel is W (l) , with a bias of b (l) , the activation function is σ, then the output feature map of the lth layer is:

[0035] H (l) =σ(W (l) *H (l-1) +b (l) )

[0036] Among them, * represents the convolution operation; when training the CNN model, the output of the neuron is set to zero with probability p to alleviate overfitting;

[0037] Let the output of layer l be z (l) , the output after Dropout is but: where γ (l) is a random variable between 0 and 1, ⊙ represents the Hadamard product;

[0038] Using the cross entropy loss function, let the true label be y and the predicted probability distribution be but:

[0039]

[0040] in, is the predicted probability distribution of the j-th material category.

[0041] Preferably, the waste lithium battery particles after the three-stage grinding and crushing are placed on a conveyor belt, and the spectral data and element information data of the waste lithium battery particles are collected in real time using an online spectrometer and an X-ray fluorescence spectrometer, and the collected data are input into a trained electrode material sorting model;

[0042] The electrode material sorting model goes through multiple convolutional layers, pooling layers, and fully connected layers to output the probability vector of the category to which the waste lithium battery particles belong;

[0043] The category with the largest median value of the probability vector is taken as the predicted category of the current crushing material; if the predicted category is electrode material, the current crushing material is determined to be electrode material particles; otherwise, it is determined to be non-electrode material particles; according to the real-time discrimination results of the CNN model, the action of the pneumatic sorter is controlled to sort the electrode material particles and non-electrode material particles into different collection devices:

[0044] When the CNN model identifies the electrode material particles, the left solenoid valve of the pneumatic sorter is opened to blow the electrode material particles into a collection box on the side of the conveyor belt in the forward direction;

[0045] When the CNN model identifies non-electrode material particles, the right solenoid valve of the pneumatic sorter is opened to blow the non-electrode material particles into the collection box on the other side of the conveyor belt's forward direction;

[0046] The non-electrode material particles collected after pneumatic sorting are subjected to multi-frequency vibration screening; the multi-frequency vibration screening is performed by a multi-frequency vibration screening device, wherein the multi-frequency vibration screening device is composed of a vibrating screen, an eccentric motor and a spring shock absorber, and the vibration frequency of the screen is controlled by adjusting the rotation speed of the eccentric motor;

[0047] Constructing a multi-frequency vibration screening model:

[0048]

[0049] Among them, k is the stiffness coefficient of the spring shock absorber, f is the vibration frequency, and m is the screen load mass; impurity particles with a density of ρ2 and an average particle size of R2 will pass through the screen with a mesh size of d, and electrode material particles with a density of ρ1 and an average particle size of R1 will be retained by the screen;

[0050] By controlling the speed of the eccentric motor, the screen vibration of different frequencies is generated. The frequencies from low to high are f1, f2, ..., f n , and satisfy f1 <f2<...<f n ; Perform vibration screening at frequency f1, and then gradually increase the vibration frequency to f n , to achieve the stratification of various materials in non-electrode material particles.

[0051] Preferably, during the multi-frequency vibration screening process, the material thickness on the upper surface of the screen is monitored; a laser distance sensor is arranged above the screen to measure the distance between the upper surface of the screen and the sensor, and the thickness of the material layer is calculated according to the structural dimensions of the multi-frequency vibration screening device;

[0052] Assume the length of the screen is L S , width W s , the initial distance between the screen and the sensor is H0, and the distance measured at time κ is H κ , then the material layer thickness h at time κ κ For: h κ =H0-H κ ;

[0053] By adjusting the material feed rate v in and vibration frequency f, the thickness of the material layer is controlled within [hmin ,h max ]; where the feed rate v in The following relationship is satisfied with the vibration frequency f:

[0054] f∈[f min ,f max ],h κ ∈[h min ,h max ]

[0055] Among them, m t is the mass of feed per unit time, ρ d is the packing density of electrode material particles, f min and f max are the minimum and maximum values ​​of the vibration frequency.

[0056] Preferably, after collecting the electrode material particles, electrostatic separation technology is used to separate the positive electrode material and the negative electrode material particles in the electrode material particles; the collected electrode material particles are evenly spread on a grounded conveyor belt;

[0057] A high-voltage static electrode is arranged above the conveyor belt, and a positive electrode material recovery box and a negative electrode material recovery box are arranged below the high-voltage static electrode, and a partition is arranged between the two recovery boxes; a strong electric field is formed between the high-voltage static electrode and the grounded conveyor belt by applying a DC high voltage to the high-voltage static electrode; when the electrode material particles enter the electric field, the negative electrode material with good conductivity is quickly charged, and is subjected to a large electrostatic force, and is attracted and deflected into the negative electrode material recovery box; while the positive electrode material with relatively poor conductivity is subjected to a small electrostatic force, maintains the original motion trajectory, and enters the positive electrode material recovery box;

[0058] The DC voltage applied to the high voltage static electrode is U e , the distance between the static electrode and the conveyor belt is d ec The average particle size of the electrode material particles is The charge is q, and when the following equation is satisfied, the negative electrode material particles are deflected and sorted:

[0059]

[0060] Among them, M k is the mass of the particle, g is the acceleration due to gravity, μ is the friction factor between the particle and the conveyor belt, L e is the length of the static electrode; by adjusting the applied voltage U e The size of the particle controls the sensitivity of the sorting process to the conductivity of the particles.

[0061] A waste lithium battery recycling system based on multi-stage crushing and intelligent sorting, which is used to implement the waste lithium battery recycling method based on multi-stage crushing and intelligent sorting, comprising: a discharge processing module, a multi-stage crushing module, a pneumatic sorting module, a non-electrode material separation module and an electrode material separation module;

[0062] The discharge processing module is used to perform immersion discharge processing on the waste lithium batteries to be recycled, build a waste lithium battery discharge detection model, and detect whether the waste lithium batteries have completed discharge;

[0063] The multi-stage crushing module is used to perform multi-stage crushing on the lithium battery after the discharge treatment, including primary shear crushing, secondary impact crushing and tertiary grinding crushing. In the graded crushing process, the particle size of the lithium battery is graded and screened;

[0064] The pneumatic sorting module is used to obtain the spectrum and element information of the crushed lithium battery material, build an electrode material sorting model based on the deep learning model, identify the electrode material based on the spectrum and element information of the crushed lithium battery material, and control the pneumatic sorter to sort the electrode material and non-electrode material;

[0065] The non-electrode material separation module collects the lithium battery electrode material particles and the non-electrode material particles after pneumatic sorting, applies multi-frequency vibration to the collected non-electrode material particles, and separates the non-electrode material particles;

[0066] The electrode material separation module performs electrostatic separation on the collected electrode material particles to separate the positive electrode material and the negative electrode material in the electrode material particles.

[0067] Beneficial effects of the present invention: The present invention ensures that the waste lithium battery is completely discharged and eliminates potential safety hazards by immersion discharge and construction of a detection model. After discharge, the battery is cleaned and dried to remove impurities and prepare for subsequent recycling.

[0068] The present invention adopts multi-stage crushing and graded screening to crush the waste lithium battery materials to an ideal particle size and achieve safety and controllability; it ensures uniform crushing and is conducive to efficient material sorting and recycling.

[0069] The present invention utilizes spectral element information and deep learning models to intelligently identify electrode materials, and combines pneumatic sorting to quickly and accurately separate electrode and non-electrode materials, thereby overcoming the defects of manual sorting and improving efficiency.

[0070] The present invention applies multi-frequency vibration to non-electrode materials for layered separation, adjusts the vibration frequency according to the difference in material properties to achieve gradient separation, monitors and optimizes process parameters, ensures separation effect, and improves efficiency.

[0071] The present invention uses electrostatic separation to separate the positive and negative electrodes of electrode materials. Based on the difference in conductivity, accurate and efficient separation is achieved by adjusting the electric field, laying the foundation for the purification and recovery of electrode materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 A flow chart of a waste lithium battery recycling method based on multi-stage crushing and intelligent sorting provided by the present invention;

[0073] Figure 2 This is a structural diagram of the waste lithium battery recycling system based on multi-stage crushing and intelligent sorting provided by the present invention. DETAILED DESCRIPTION

[0074] In order to better understand the present invention, a more detailed description will be made of various aspects of the present invention with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0075] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The accompanying drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in the present invention, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0076] It should also be understood that expressions such as "comprises", "including", "having", "includes" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present invention, "may" is used to mean "one or more embodiments of the present invention". And, the term "exemplary" is intended to refer to an example or illustration.

[0077] Unless otherwise defined, all words (including engineering terms and scientific and technological terms) used in this article have the same meaning as those commonly understood by ordinary technicians in the field to which the present invention belongs. It should also be understood that unless there is a clear explanation in the present invention, the words defined in the commonly used dictionary should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0078] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0079] Example 1

[0080] Reference Figure 1 , which is the first embodiment of the present invention, provides a waste lithium battery recycling method based on multi-stage crushing and intelligent sorting.

[0081] S1: The waste lithium batteries to be recycled are subjected to immersion discharge treatment, and a waste lithium battery discharge detection model is constructed to detect whether the waste lithium batteries have completed discharge.

[0082] The waste lithium battery is immersed in a salt solution tank for discharge. A current sensor is connected to the total output circuit of the battery to monitor the change of the total output current I(t) in real time. The discharge detection model of the waste lithium battery is:

[0083]

[0084] Where I(t) represents the total output current of the battery pack at time t, n is the number of used lithium batteries, and I i (t) represents the output current of the ith battery at time t, U i represents the initial voltage of the ith battery, i.e., the open circuit voltage before immersion discharge, R i represents the internal resistance of the ith battery, t represents the immersion discharge time, τ i represents the time constant of the ith battery, τ i =R i ×C i , where C i is the capacity of the ith battery.

[0085] The physical meaning of the waste lithium battery discharge detection model is: during the immersion discharge process, the output current of each battery will gradually decrease in an exponential decay manner until it approaches zero. The initial voltage U i and internal resistance R i determine the initial current value, and the time constant τ i Determines the rate at which the current decays.

[0086] The total output current I(t) of the battery pack is the superposition of the output currents of all batteries. By detecting the size of I(t), the discharge process of the used lithium battery can be judged.

[0087] Set the discharge current threshold I th , when the measured total output current I(t) is less than or equal to I th , it is determined that all batteries have been discharged.

[0088] After the waste lithium battery is completely discharged, the lithium battery soaked in the salt solution is rinsed with water to remove the residual salt solution and impurities on the surface of the lithium battery, and then dried.

[0089] By subjecting used lithium batteries to an immersion discharge treatment and building a used lithium battery discharge detection model, the total output current changes of the battery pack are monitored in real time to accurately determine whether the used lithium batteries are completely discharged. This effectively avoids safety hazards caused by residual power and lays the foundation for subsequent disassembly and processing. At the same time, the battery is cleaned and dried with water after discharge to further remove surface impurities and improve recycling efficiency.

[0090] S2: The lithium battery after discharge treatment is subjected to multi-stage crushing, including primary shear crushing, secondary impact crushing and tertiary grinding crushing. During the graded crushing process, the lithium battery particle size is graded and screened.

[0091] The waste lithium batteries after discharge treatment are crushed, and the crushing process includes primary shear crushing, secondary impact crushing and tertiary grinding crushing.

[0092] The adaptive feeding control of waste lithium batteries based on machine vision is introduced. The convolutional neural network (CNN) is used to extract features and semantic segment the images in the crushing chamber, monitor the material level of waste lithium batteries in real time, and adjust the feeding rate by detecting the material level in the crushing chamber in real time.

[0093] The waste lithium batteries are subjected to primary shearing and crushing treatment. A temperature sensor is set in the shearing and crushing chamber of the shearing and crushing treatment, and the temperature threshold T of the shearing and crushing chamber is set. δ When the temperature T in the shear crushing chamber is detected to be greater than the temperature threshold T δ When the shearing and crushing is stopped, the shearing and crushing chamber is sprayed with water for cooling. After the water spraying and cooling is stopped, it is detected whether the temperature T in the shearing and crushing chamber is less than the temperature threshold T δ , if T <T δ , then continue shearing and crushing; if T ≥ T δ , a high temperature alarm is triggered.

[0094] Setting the particle size threshold D of waste lithium batteries during primary shear crushing p1 , and according to the crushing particle size threshold D p1Set the first-level screen mesh diameter d1, D p1 =d1; the crushed waste lithium batteries that cannot pass through the first-level screen are sheared and crushed again.

[0095] After the waste lithium batteries have completed the primary shear crushing, the crushed waste lithium batteries are subjected to secondary impact crushing.

[0096] Before impact crushing the broken waste lithium batteries, nitrogen is input into the impact crushing chamber to replace the air in the impact crushing chamber until the oxygen concentration in the impact crushing chamber is lower than the oxygen safety concentration threshold. An oxygen analyzer is arranged at the exhaust port of the impact crushing chamber to detect the oxygen concentration in the impact crushing chamber. When it is detected that the oxygen concentration in the impact crushing chamber is lower than the oxygen safety concentration threshold, the broken waste lithium batteries after the primary shear crushing are put into the impact crushing chamber for secondary impact crushing.

[0097] During the impact crushing process, nitrogen is continuously input to maintain the positive nitrogen pressure in the impact crushing chamber, and the oxygen concentration in the impact crushing chamber is continuously detected. If the oxygen concentration is greater than the oxygen safety concentration threshold, the impact crushing is stopped.

[0098] The nitrogen discharged during the crushing process is dried, dust-filtered and reused;

[0099] Setting the particle size threshold D of waste lithium batteries during secondary impact crushing p2 , and according to the crushing particle size threshold D p2 Set up the secondary screen mesh diameter d2, D p2 =d2; the crushed waste lithium batteries that cannot pass through the secondary screen are re-impacted and crushed.

[0100] The lithium battery after impact crushing is subjected to three-stage grinding and crushing to grind the crushed lithium battery into powder particles. The crushing particle size threshold D of the waste lithium battery during the three-stage grinding and crushing is set. p3 , and according to the crushing particle size threshold D p3 Set up a three-level screen mesh diameter d3, D p3 =d3; the crushed waste lithium batteries that cannot pass through the third-level screen are re-grinded and crushed until all the ground particles of the waste lithium batteries pass through the third-level screen.

[0101] A multi-stage crushing method is adopted, through the first-stage shear crushing, the second-stage impact crushing and the third-stage grinding crushing, the waste lithium battery materials are gradually crushed to a suitable particle size; machine vision technology is introduced to achieve adaptive feed control, and safety measures such as temperature monitoring and inert gas protection are taken; multi-stage screening ensures that the crushed particle size is uniform, creating favorable conditions for subsequent material sorting and improving resource recycling rate.

[0102] S3: Obtain the spectrum and element information of the crushed lithium battery materials, build an electrode material sorting model based on the deep learning model, identify the electrode materials based on the spectrum and element information of the crushed lithium battery materials, and control the pneumatic sorter to sort the electrode materials and non-electrode materials.

[0103] X-ray fluorescence spectrometer (XRF) and laser induced breakdown spectrometer (LIBS) were used to collect spectral data and element information data of N different types of lithium battery materials. M samples were collected for each lithium battery, totaling N×M samples. The lithium battery material samples were labeled into K material categories, with labels y i ∈1,2,...,K; denoise and normalize the spectral data.

[0104] Perform PCA feature extraction on the spectral data of lithium battery materials, and set the normalized spectral data matrix as The principal component matrix obtained by PCA is Then: X=U∑V T ; Where P is the number of principal components selected by PCA, ∑ is the singular value matrix, V is the right singular matrix, V T is the transpose of the right singular matrix; the spectral feature matrix after dimension reduction is:

[0105] Perform ICA feature extraction on the element information data of lithium battery electrode materials, assuming that the element information matrix is The independent component matrix obtained by ICA is Then: Y = AS; where A is the mixing matrix, S is the independent component, D is the number of element types, and Q is the number of selected independent components; the element feature matrix after dimensionality reduction is

[0106] Combine the spectral feature matrix and the element feature matrix to convert the spectral feature matrix X pca and element feature matrix Y ica Splice in the column direction to obtain a (P+Q)-dimensional sample feature matrix

[0107] Construct an electrode material sorting model based on the convolutional neural network CNN model. The input layer is a (P+Q)-dimensional sample feature matrix. After C convolutional layers, G pooling layers and B fully connected layers, the output is a K-dimensional material category probability vector.

[0108] In the CNN model, let the output feature map of the l-1th layer be H (l-1) , the convolution kernel is W (l) , with a bias of b (l) , the activation function is σ, then the output feature map of the lth layer is:

[0109] H (l) =σ(W (l) *H (l-1) +b (l) )

[0110] Among them, * represents the convolution operation; when training the CNN model, the output of the neuron is set to zero with probability p to alleviate overfitting.

[0111] Let the output of layer l be z (l) , the output after Dropout is but: where γ (l) is a random variable between 0 and 1, and ⊙ represents the Hadamard product.

[0112] Using the cross entropy loss function, let the true label be y and the predicted probability distribution be but:

[0113]

[0114] in, is the predicted probability distribution of the j-th material category.

[0115] The waste lithium battery particles after three-stage grinding and crushing are placed on a conveyor belt, and the spectral data and elemental information data of the waste lithium battery particles are collected in real time using an online spectrometer and X-ray fluorescence spectrometer, and the collected data are input into the trained electrode material sorting model.

[0116] The electrode material sorting model goes through multiple convolutional layers, pooling layers and fully connected layers to output the probability vector of the category to which the waste lithium battery particles belong.

[0117] The category with the largest median value of the probability vector is taken as the predicted category of the current crushing material; if the predicted category is electrode material, the current crushing material is determined to be electrode material particles; otherwise, it is determined to be non-electrode material particles; according to the real-time discrimination results of the CNN model, the action of the pneumatic sorter is controlled to sort the electrode material particles and non-electrode material particles into different collection devices;

[0118] When the CNN model identifies the electrode material particles, the left solenoid valve of the pneumatic sorter is opened to blow the electrode material particles into a collection box on the side of the conveyor belt in the forward direction;

[0119] When the CNN model identifies non-electrode material particles, the right solenoid valve of the pneumatic sorter is opened to blow the non-electrode material particles into the collection box on the other side of the conveyor belt's forward direction.

[0120] XRF and LIBS are used to collect the spectrum and elemental information of lithium battery materials, and a CNN deep learning model is built to intelligently identify electrode materials. Through pneumatic sorter control, electrode materials are quickly and accurately separated from non-electrode materials. The defects of traditional manual sorting such as low efficiency and poor accuracy are overcome, and efficient and accurate recycling of electrode materials is achieved.

[0121] S4: Collect the lithium battery electrode material particles and non-electrode material particles after pneumatic sorting respectively, apply multi-frequency vibration to the collected non-electrode material particles, and stratify the non-electrode material particles.

[0122] The non-electrode material particles collected after pneumatic sorting are subjected to multi-frequency vibration screening; the multi-frequency vibration screening is performed by a multi-frequency vibration screening device, which is composed of a vibrating screen, an eccentric motor and a spring shock absorber, and the vibration frequency of the screen is controlled by adjusting the rotation speed of the eccentric motor.

[0123] Constructing a multi-frequency vibration screening model:

[0124]

[0125] Among them, k is the stiffness coefficient of the spring shock absorber, f is the vibration frequency, and m is the load mass of the screen; impurity particles with a density of v2 and an average particle size of R2 will pass through the screen with a mesh size of d, and electrode material particles with a density of ρ1 and an average particle size of R1 will be retained by the screen.

[0126] By utilizing the differences in density (ρ), elasticity, damping and other properties of each material in the non-electrode material, and applying vibrations of a specific frequency (f), these different material particles are gradually layered on the sieve, thereby achieving effective separation of each non-electrode material particle.

[0127] By controlling the speed of the eccentric motor, the screen vibration of different frequencies is generated. The frequencies from low to high are f1, f2, ..., f n , and satisfy f1 <f2<...<f n ; Perform vibration screening at frequency f1 to quickly separate non-electrode particles with large density differences, and then gradually increase the vibration frequency to f n , so that non-electrode particles with smaller separation density differences can be separated, and various materials in non-electrode material particles can be separated.

[0128] During the multi-frequency vibration screening process, the material thickness on the surface of the screen is monitored; a laser distance sensor is set above the screen to measure the distance between the surface of the screen and the sensor, and the thickness of the material layer is calculated based on the structural dimensions of the multi-frequency vibration screening device.

[0129] Assume the length of the screen is L s , width Ws , the initial distance between the screen and the sensor is H0, and the distance measured at time κ is H κ , then the material layer thickness h at time κ κ For: h κ =H0-H κ .

[0130] By adjusting the material feed rate v in and vibration frequency f, the thickness of the material layer is controlled within [h min ,h max ]; this ensures that the material has enough loose space during screening, avoids too long screening time, and improves screening efficiency. Among them, the feed rate v in The following relationship is satisfied with the vibration frequency f:

[0131] f∈[f min ,f max ],h κ ∈[h min ,h max ]

[0132] Among them, m t is the mass of feed per unit time, ρ d is the packing density of electrode material particles, f min and f max are the minimum and maximum values ​​of the vibration frequency.

[0133] Multi-frequency vibration screening is applied to the non-electrode material particles after pneumatic sorting. According to the differences in density, particle size and other properties of different materials, the gradient stratification separation of the materials is achieved by adjusting the vibration frequency. A laser ranging sensor is introduced to monitor the material thickness on the screen in real time. By optimizing the feed rate and vibration frequency, the screening time is shortened while ensuring the sorting effect, which significantly improves the sorting efficiency of non-electrode materials.

[0134] S5: electrostatically sorting the collected electrode material particles to separate the positive electrode material and the negative electrode material in the electrode material particles.

[0135] After collecting the electrode material particles, electrostatic separation technology is used to separate the positive electrode material and negative electrode material particles in the electrode material particles. Electrostatic separation is a method of material separation that utilizes the difference in conductivity of different materials and the difference in electrostatic force they experience in an external electric field.

[0136] The collected electrode material particles are evenly spread on a grounded conveyor belt, a high-voltage static electrode is arranged above the conveyor belt, a positive electrode material recovery box and a negative electrode material recovery box are arranged below the conveyor belt, and a partition is arranged between the two recovery boxes; a strong electric field is formed between the high-voltage static electrode and the grounded conveyor belt by applying DC high voltage to the high-voltage static electrode; when the electrode material particles enter the electric field, the negative electrode material with good conductivity is quickly charged, and is affected by a large electrostatic force, and is attracted and deflected into the negative electrode material recovery box; while the positive electrode material with relatively poor conductivity is affected by a small electrostatic force, maintains its original motion trajectory, and enters the positive electrode material recovery box.

[0137] The DC voltage applied to the high voltage static electrode is U e , the distance between the static electrode and the conveyor belt is d ec The average particle size of the electrode material particles is The charge is q, the electric field strength is E, then the electrostatic force F on the particle in the electric field is e satisfy:

[0138]

[0139] When the following equation is satisfied, the negative electrode material particles with good conductivity are effectively deflected and sorted:

[0140]

[0141] Among them, M k is the mass of the particle, g is the acceleration due to gravity, μ is the friction factor between the particle and the conveyor belt, L e is the length of the static electrode; by adjusting the applied voltage U e The size of the particle controls the sensitivity of the sorting process to the conductivity of the particles.

[0142] Electrostatic sorting technology is used to separate positive and negative electrode materials from electrode material particles; based on the difference in conductivity between the two materials, the high-voltage electrostatic field deflects the negative electrode material with good conductivity into the negative electrode recovery box, while the positive electrode material with relatively poor conductivity maintains its original trajectory and enters the positive electrode recovery box. By adjusting the voltage to control the sorting sensitivity, efficient and accurate separation of positive and negative electrode materials is achieved, laying the foundation for the purification of electrode materials.

[0143] Example 2

[0144] Reference Figure 2 , which is the second embodiment of the present invention, provides a waste lithium battery recycling system based on multi-stage crushing and intelligent sorting.

[0145] The system comprises a discharge processing module, a multi-stage crushing module, a pneumatic sorting module, a non-electrode material separation module and an electrode material separation module.

[0146] The discharge processing module is used to perform immersion discharge processing on the waste lithium batteries to be recycled, build a waste lithium battery discharge detection model, and detect whether the waste lithium batteries have completed discharge.

[0147] The multi-stage crushing module is used to perform multi-stage crushing on the lithium battery after discharge treatment, including primary shear crushing, secondary impact crushing and tertiary grinding crushing. In the graded crushing process, the lithium battery particle size is graded and screened.

[0148] The pneumatic sorting module is used to obtain the spectrum and element information of the crushed lithium battery material, build an electrode material sorting model based on the deep learning model, identify the electrode material based on the spectrum and element information of the crushed lithium battery material, and control the pneumatic sorter to sort the electrode material and non-electrode material;

[0149] The non-electrode material separation module collects the lithium battery electrode material particles and the non-electrode material particles after pneumatic sorting, applies multi-frequency vibration to the collected non-electrode material particles, and separates the non-electrode material particles;

[0150] The electrode material separation module performs electrostatic separation on the collected electrode material particles to separate the positive electrode material and the negative electrode material in the electrode material particles.

[0151] Example 3

[0152] The present invention also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the waste lithium battery recycling method based on multi-stage crushing and intelligent sorting as described above may be executed.

[0153] The method or system according to the embodiment of the present invention may also be implemented with the aid of the architecture of the electronic device of the present invention.

[0154] An electronic device may include a bus, one or more CPUs, read-only memory (ROM), random access memory (RAM), a communication port for connecting to a network, input / output components, a hard disk, and the like.

[0155] The storage device in the electronic device, such as ROM or hard disk, can store the waste lithium battery recycling method based on multi-stage crushing and intelligent sorting provided by the present invention.

[0156] A waste lithium battery recycling method based on multi-stage crushing and intelligent sorting includes: immersing and discharging the waste lithium batteries to be recycled, constructing a waste lithium battery discharge detection model, and detecting whether the waste lithium batteries have completed discharge; multi-stage crushing of the lithium batteries after discharge treatment, including primary shear crushing, secondary impact crushing, and tertiary grinding crushing, and in the graded crushing process, graded screening and crushing the lithium battery particle size; obtaining the spectrum and element information of the crushed lithium battery materials, constructing an electrode material sorting model based on a deep learning model, identifying the electrode material based on the spectrum and element information of the crushed lithium battery materials, and controlling a pneumatic sorter to sort the electrode material and non-electrode material; respectively collecting the lithium battery electrode material particles and non-electrode material particles after pneumatic sorting, applying multi-frequency vibration to the collected non-electrode material particles, and stratifying the non-electrode material particles; electrostatically sorting the collected electrode material particles to separate the positive electrode material and the negative electrode material in the electrode material particles.

[0157] Furthermore, the electronic device may also include a user interface. Of course, the architecture of the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0158] Example 4

[0159] The invention also discloses a computer-readable storage medium.

[0160] The computer-readable storage medium has computer-readable instructions stored thereon.

[0161] When the computer-readable instructions are executed by the processor, the waste lithium battery recycling method based on multi-stage crushing and intelligent sorting according to the embodiment of the present invention described with reference to the above drawings can be executed.

[0162] The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. In addition, according to an embodiment of the present invention, the process described above with reference to the flowchart may be implemented as a computer software program.

[0163] For example, the present invention provides a non-temporary machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided by the present invention, for example: performing immersion discharge treatment on the recycled waste lithium batteries, constructing a waste lithium battery discharge detection model, and detecting whether the waste lithium batteries have completed discharge; performing multi-stage crushing on the lithium batteries after discharge treatment, including primary shear crushing, secondary impact crushing, and tertiary grinding crushing, and in the graded crushing process, graded screening and crushing the lithium battery particle size; obtaining the spectrum and element information of the crushed lithium battery materials, constructing an electrode material sorting model based on a deep learning model, identifying the electrode material based on the spectrum and element information of the crushed lithium battery materials, and controlling the pneumatic sorter to sort the electrode material and non-electrode material; respectively collecting the lithium battery electrode material particles and non-electrode material particles after pneumatic sorting, applying multi-frequency vibration to the collected non-electrode material particles, and stratifying the non-electrode material particles; electrostatically sorting the collected electrode material particles to separate the positive electrode material and the negative electrode material in the electrode material particles.

[0164] When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present invention are performed. The method, apparatus, and device of the present invention may be implemented in many ways. For example, the method, apparatus, and device of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware.

[0165] The above sequence for the steps of the method is for illustration only, and the steps of the method of the present invention are not limited to the sequence specifically described above unless otherwise specifically stated.

[0166] In addition, in some embodiments, the present invention can also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Therefore, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0167] In addition, the parts of the above technical solutions provided in the embodiments of the present invention that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0168] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A waste lithium battery recycling method based on multi-stage crushing and intelligent sorting, characterized in that: include: The recycled waste lithium batteries are subjected to immersion discharge treatment, and a waste lithium battery discharge detection model is constructed to detect whether the waste lithium batteries have completed discharge; The lithium battery after discharge treatment is crushed in multiple stages, including primary shear crushing, secondary impact crushing and tertiary grinding crushing. In the process of graded crushing, the particle size of lithium battery is graded and sieved; Obtain the spectrum and element information of the crushed lithium battery materials, build an electrode material sorting model based on the deep learning model, identify the electrode materials based on the spectrum and element information of the crushed lithium battery materials, and control the pneumatic sorter to sort the electrode materials and non-electrode materials; Separately collect the lithium battery electrode material particles and non-electrode material particles after pneumatic sorting, apply multi-frequency vibration to the collected non-electrode material particles, and stratify the non-electrode material particles; The collected electrode material particles are subjected to electrostatic separation to separate the positive electrode material and the negative electrode material in the electrode material particles.

2. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 1 is characterized in that: The waste lithium battery is immersed in a salt solution tank for discharge. A current sensor is connected to the total output circuit of the battery to monitor the change of the total output current I(t) in real time. The discharge detection model of the waste lithium battery is: Where I(t) represents the total output current of the battery pack at time t, n is the number of used lithium batteries, and I i (t) represents the output current of the ith battery at time t, U i represents the initial voltage of the ith battery, i.e., the open circuit voltage before immersion discharge, R i represents the internal resistance of the ith battery, t represents the immersion discharge time, τ i represents the time constant of the ith battery, τ i =R i ×C i , where C i is the capacity of the i-th battery; The total output current I(t) of the battery pack is the superposition of the output currents of all batteries. By detecting the size of I(t), the discharge process of the waste lithium battery can be judged; Set the discharge current threshold I th , when the measured total output current I(t) is less than or equal to I th When , it is determined that all batteries have been discharged; After the waste lithium battery is completely discharged, the lithium battery soaked in the salt solution is rinsed with water to remove the residual salt solution and impurities on the surface of the lithium battery, and then dried.

3. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 2 is characterized in that: The waste lithium batteries after the discharge treatment are crushed, and the crushing process includes primary shear crushing, secondary impact crushing and tertiary grinding crushing; Introducing adaptive feeding control of waste lithium batteries based on machine vision, using convolutional neural network (CNN) to extract features and segment semantics of images in the crushing chamber, monitoring the material level of waste lithium batteries in real time, and adjusting the feeding rate by real-time detection of the material level in the crushing chamber; The waste lithium batteries are subjected to primary shearing and crushing treatment. A temperature sensor is set in the shearing and crushing chamber of the shearing and crushing treatment, and the temperature threshold T of the shearing and crushing chamber is set. δ When the temperature T in the shear crushing chamber is detected to be greater than the temperature threshold T δ When the shearing and crushing is stopped, the shearing and crushing chamber is sprayed with water for cooling. After the water spraying and cooling is stopped, it is detected whether the temperature T in the shearing and crushing chamber is less than the temperature threshold T δ , if T <T δ , then continue shearing and crushing; if T ≥ T δ , then the high temperature alarm is triggered; Setting the particle size threshold D of waste lithium batteries during primary shear crushing p1 , and according to the crushing particle size threshold D p1 Set the first-level screen mesh diameter d1, D p1 =d1; the crushed waste lithium batteries that cannot pass through the first-level screen are sheared and crushed again.

4. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 3 is characterized in that: After the waste lithium batteries have completed primary shear crushing, the crushed waste lithium batteries are subjected to secondary impact crushing; Before impact crushing the broken waste lithium batteries, nitrogen is input into the impact crushing chamber to replace the air in the impact crushing chamber by the input nitrogen until the oxygen concentration in the impact crushing chamber is lower than the oxygen safety concentration threshold, an oxygen analyzer is arranged at the exhaust port of the impact crushing chamber to detect the oxygen concentration in the impact crushing chamber, and when it is detected that the oxygen concentration in the impact crushing chamber is lower than the oxygen safety concentration threshold, the broken waste lithium batteries after primary shear crushing are put into the impact crushing chamber for secondary impact crushing; During the impact crushing process, nitrogen is continuously input to maintain the positive nitrogen pressure in the impact crushing chamber, and the oxygen concentration in the impact crushing chamber is continuously detected. If the oxygen concentration is greater than the oxygen safety concentration threshold, the impact crushing is stopped; The nitrogen discharged during the crushing process is dried, dust-filtered and reused; Setting the particle size threshold D of waste lithium batteries during secondary impact crushing p2 , and according to the crushing particle size threshold D p2 Set up the secondary screen mesh diameter d2, D p2 =d2; re-impact and crush the broken waste lithium batteries that cannot pass through the secondary screen; The lithium battery after impact crushing is subjected to three-stage grinding and crushing to grind the crushed lithium battery into powder particles. The crushing particle size threshold D of the waste lithium battery during the three-stage grinding and crushing is set. p3 , and according to the crushing particle size threshold D p3 Set up a three-level screen mesh diameter d3, D p3 =d3; the crushed waste lithium batteries that cannot pass through the third-level screen are re-grinded and crushed until all the ground particles of the waste lithium batteries pass through the third-level screen.

5. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 4 is characterized in that: X-ray fluorescence spectrometer and laser induced breakdown spectrometer are used to collect spectral data and element information data of N different types of lithium battery materials. M samples are collected for each lithium battery, totaling N×M samples. The lithium battery material samples are labeled into K material categories, with labels y i ∈1,2,...,K; denoise and normalize the spectral data; Perform PCA feature extraction on the spectral data of lithium battery materials, and assume that the normalized spectral data matrix is The principal component matrix obtained by PCA is Then: X=U∑V T ; Where P is the number of principal components selected by PCA, ∑ is the singular value matrix, V is the right singular matrix, V T is the transpose of the right singular matrix; the spectral feature matrix after dimension reduction is: Perform ICA feature extraction on the element information data of lithium battery materials, assuming that the element information matrix is The independent component matrix obtained by ICA is Then: Y = AS; where A is the mixing matrix, S is the independent component, D is the number of element types, and Q is the number of selected independent components; the element feature matrix after dimensionality reduction is: Combine the spectral feature matrix and the element feature matrix to convert the spectral feature matrix X pca and element feature matrix Y ica Splice in the column direction to obtain a (P+Q)-dimensional sample feature matrix 6. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 5 is characterized in that: Construct an electrode material sorting model based on the convolutional neural network CNN model. The input layer is a (P+Q)-dimensional sample feature matrix. After C convolutional layers, G pooling layers and B fully connected layers, the output is a K-dimensional material category probability vector. In the CNN model, let the output feature map of the l-1th layer be H (l-1) , the convolution kernel is W (l) , with a bias of b (l) , the activation function is σ, then the output feature map of the lth layer is: H (l) =σ(W (l) *H (l-1) +b (l) ) Among them, * represents the convolution operation; when training the CNN model, the output of the neuron is set to zero with probability p to alleviate overfitting; Let the output of layer l be z (l) , the output after Dropout is but: where γ (l) is a random variable between 0 and 1, ⊙ represents the Hadamard product; Using the cross entropy loss function, let the true label be y and the predicted probability distribution be but: in, is the predicted probability distribution of the j-th material category.

7. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 6 is characterized in that: The waste lithium battery particles after three-stage grinding and crushing are placed on a conveyor belt, and the spectral data and element information data of the waste lithium battery particles are collected in real time using an online spectrometer and an X-ray fluorescence spectrometer, and the collected data are input into the trained electrode material sorting model; The electrode material sorting model goes through multiple convolutional layers, pooling layers, and fully connected layers to output the probability vector of the category to which the waste lithium battery particles belong; The category with the largest median value of the probability vector is taken as the predicted category of the current crushing material; if the predicted category is electrode material, the current crushing material is determined to be electrode material particles; otherwise, it is determined to be non-electrode material particles; according to the real-time discrimination results of the CNN model, the action of the pneumatic sorter is controlled to sort the electrode material particles and non-electrode material particles into different collection devices: When the CNN model identifies the electrode material particles, the left solenoid valve of the pneumatic sorter is opened to blow the electrode material particles into a collection box on the side of the conveyor belt in the forward direction; When the CNN model identifies non-electrode material particles, the right solenoid valve of the pneumatic sorter is opened to blow the non-electrode material particles into the collection box on the other side of the conveyor belt's forward direction; The non-electrode material particles collected after pneumatic sorting are subjected to multi-frequency vibration screening; the multi-frequency vibration screening is performed by a multi-frequency vibration screening device, wherein the multi-frequency vibration screening device is composed of a vibrating screen, an eccentric motor and a spring shock absorber, and the vibration frequency of the screen is controlled by adjusting the rotation speed of the eccentric motor; Constructing a multi-frequency vibration screening model: Among them, k is the stiffness coefficient of the spring shock absorber, f is the vibration frequency, and m is the screen load mass; impurity particles with a density of ρ2 and an average particle size of R2 will pass through the screen with a mesh size of d, and electrode material particles with a density of ρ1 and an average particle size of R1 will be retained by the screen; By controlling the speed of the eccentric motor, the screen vibration of different frequencies is generated. The frequencies from low to high are f1, f2, ..., f n , and satisfy f1 <f2<...<f n ; Perform vibration screening at frequency f1, and then gradually increase the vibration frequency to f n , to achieve the stratification of various materials in non-electrode material particles.

8. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 7 is characterized in that: During the multi-frequency vibration screening process, the material thickness on the surface of the screen is monitored; a laser distance sensor is set above the screen to measure the distance between the surface of the screen and the sensor, and the thickness of the material layer is calculated based on the structural dimensions of the multi-frequency vibration screening device; Assume the length of the screen is L S , width W s , the initial distance between the screen and the sensor is H0, and the distance measured at time κ is H κ , then the material layer thickness h at time κ κ For: h κ =H0-H κ ; By adjusting the material feed rate v in and vibration frequency f, the thickness of the material layer is controlled within [h min ,h max ]; where the feed rate v in The following relationship is satisfied with the vibration frequency f: Among them, m t is the mass of feed per unit time, ρ d is the packing density of electrode material particles, f min and f max are the minimum and maximum values ​​of the vibration frequency.

9. The method for recycling waste lithium batteries based on multi-stage crushing and intelligent sorting according to claim 8, characterized in that: After collecting the electrode material particles, the positive electrode material and the negative electrode material particles in the electrode material particles are separated by electrostatic separation technology; the collected electrode material particles are evenly spread on a grounded conveyor belt; A high-voltage static electrode is arranged above the conveyor belt, and a positive electrode material recovery box and a negative electrode material recovery box are arranged below the high-voltage static electrode, and a partition is arranged between the two recovery boxes; a strong electric field is formed between the high-voltage static electrode and the grounded conveyor belt by applying a DC high voltage to the high-voltage static electrode; when the electrode material particles enter the electric field, the negative electrode material with good conductivity is quickly charged, and is subjected to a large electrostatic force, and is attracted and deflected into the negative electrode material recovery box; while the positive electrode material with relatively poor conductivity is subjected to a small electrostatic force, maintains the original motion trajectory, and enters the positive electrode material recovery box; The DC voltage applied to the high voltage static electrode is U e , the distance between the static electrode and the conveyor belt is d ec The average particle size of the electrode material particles is The charge is q, and when the following equation is satisfied, the negative electrode material particles are deflected and sorted: Among them, M k is the mass of the particle, g is the acceleration due to gravity, μ is the friction factor between the particle and the conveyor belt, L e is the length of the static electrode; by adjusting the applied voltage U e The size of the particle controls the sensitivity of the sorting process to the conductivity of the particles.

10. A waste lithium battery recycling system based on multi-stage crushing and intelligent sorting, which is used to implement the waste lithium battery recycling method based on multi-stage crushing and intelligent sorting according to any one of claims 1 to 9, characterized in that: include: Discharge processing module, multi-stage crushing module, pneumatic sorting module, non-electrode material separation module and electrode material separation module; The discharge processing module is used to perform immersion discharge processing on the waste lithium batteries to be recycled, build a waste lithium battery discharge detection model, and detect whether the waste lithium batteries have completed discharge; The multi-stage crushing module is used to perform multi-stage crushing on the lithium battery after the discharge treatment, including primary shear crushing, secondary impact crushing and tertiary grinding crushing. In the process of graded crushing, the particle size of the lithium battery is graded and screened; The pneumatic sorting module is used to obtain the spectrum and element information of the crushed lithium battery material, build an electrode material sorting model based on the deep learning model, identify the electrode material based on the spectrum and element information of the crushed lithium battery material, and control the pneumatic sorter to sort the electrode material and non-electrode material; The non-electrode material separation module collects the lithium battery electrode material particles and the non-electrode material particles after pneumatic sorting, applies multi-frequency vibration to the collected non-electrode material particles, and separates the non-electrode material particles; The electrode material separation module performs electrostatic separation on the collected electrode material particles to separate the positive electrode material and the negative electrode material in the electrode material particles.

Images